SWR LA-15 very hot

[Janou]

Hello

I have a strange problem with a SWR LA-15 2004" ( normaly the first version )

I don't know well this amplifier, just have schematic and try to understain why the plate with all electronic heat a lot. After a very short time, if i play or not, the temperature of the internal radiator is near 60°C ! ( maybe 0.3°C/s with a thermocouple K in the radiator )

Maybe it's normal, but i test several amplifier and have no problem normaly. SWR don't build Class A amp ?

I check voltages, transformer, unsolder lots of componants, check all ceramic resistors, caps, and found nothing special, no faulty componants, no short problem with transistors...

I check datasheet of transistors ( 2SC2837 ) and see eatch one dissipate 100w of total heating dissipation, maybe the radiator is very small, and everythink is normal. Maybe i need to make a correct biais ( i have no information for this one )

Nobody have this amplifier with same "problem" ?

Thanks a lot for the help.

[Mars Amp Repair]

hello,
the bias adjustment on this amp is very touchy. they want you to connect it to a 2ohm load & feed it a 1KHZ signal set so the output at the load is about 2vp-p. Then you view it on a scope & adjust the bias pot until just when the crossover distortion disappears.
you really can get it fairly close by measuring with a volt meter across 2 of the emitter resistors & setting the bias for 8-15mv (roughly). You just don't want it to be much over about 20Mv.
I dont have the schemo in front of me for the ref numbers...glen

[Mars Amp Repair]

Say,
does your amp have a little add-on circuit board at the edge of the main pcb? If it does I'd be very interested in having a close up pic of that. I don't suppose you could take a pic of the underside as it is soldered in.

It is what they call a 'noise suppression' circuit. It is designed to mute the audio drive to the output stage at power up. Perhaps just a pic of the topside might help me to build one....they are no longer available. Thanx, glen

[Enzo]

Unless there is a bad part, your bias is out of adjustment. Glen suggested one way to set it. here is another.

Monitor the mains current. Somehow connect an amp meter in series with the mains power cord. Or remove the main fuse and connect an ammeter in its place. Turn the amp on with no speaker and no signal input. The amp should draw maybe 20 watts I'd say - less than half an amp at 120VAC and less than 1/4 amp at 240VAC. If it is drawing more current than that, adjust the bias control on the power amp board. Turn it the direction that reduces the mains current. Watch as the current drops. You will reach a point where it stops reducing. That is the proper point to leave it. I would turn it just a tiny bit lower to have some margin. No one will hear a tiny bit of crossover distortion.

[Janou]

Thanks a lot for the help

I try this method this afternoon.

@ Mars Amp Repair : I read the other post, and post somes pictures to help you in few minutes

[Janou]

I'm not lucky today

I test with Enzo method, seems good, and find near 200 mA in 230v with no speaker/piezzo, and no input cable.

I try to move the biais at each extrems, and i cannot change value of drawn current. After that i set the biais at the center ( ... ) and wait 10 second to see if it raise or not. After that the drawn current raise at 240 mA, and C8 at the center of pcb blow up

[Janou]

Somes news

I change C8 and check power supply. I find +59/-59v and other supply points are OK.

@ Mars Amp Repair : R4+R3 are the emitter resistors ? It's strange because when i check in parallel, i find 0v AC or DC

I test all transistor but not Q1-Q2-Q3 ( all in radiator ) all other component are already tested, i remove all resistor or cap, transistors and all other things are OK

Any idea for the function of C8 ?

Maybe all work correclty, the only "complain" is about the temperature of metal plate ( 50°C ) and if i use the amplifier or not, when it power on, it's hot pretty quickly.

Thanks

[Enzo]

Trimmer pot VR1 is the bias control. I think you need to adjust it. I think counter clockwise. COnnect an AC ammeter in series with your mains and monitor the current. Turn VR1 so that current reduces.

[52 Bill]

...I test all transistor but not Q1-Q2-Q3 ( all in radiator ) all other component are already tested, i remove all resistor or cap, transistors and all other things are OK...

Q2 is part of the bias circuit, you should check it to see if it has failed.

Also check Q1 and Q3 for shorts or broken leads.

Because of the way that the board is mounted to the heatsink, with the transistor leads, these leads often break off.

[Janou]

Thanks

I check Q1-Q2-Q3 and i think there are OK.

I really don't know why C8 blow up... but i mesure 115VDC accross ( more than 63v and i don't understaind why the voltage is as high ) it and stop just after the power, maybe an error of mesure i must check tomorrow

I test before the method of Enzo, and find the same result, 0.24A during 2-3 seconds, decrease to 0.18A in 230v ( 41w ). I don't test all presset of VR1, because i pretty sure that C8 can blow easily.

When i open the amplifier the first time, the resistor biais is set to VR1+R8=470 Ohms, i can set to 1470 Ohms, what is the value to have the smallest global current please ?

I repair a lot of apparatus all days long, but this amplifier is strange, like i forget the most important thing

[Janou]

After a new complet check up, i find a problem with a wire...

I adjust the biais with Enzo method, and everythink is good now

Thanks a lot

[kdawg]

SWR LA-15 speaker blowing problem

Hello - I have a similar problem on an SWR LA-15, except this one ended up smoking the subwoofer. After that happened, I checked voltage on the OUT and there is 25V DC across the speaker!

Can someone help me debug this? I can see that the bases of Q3 and Q1 have voltages of 20V and -63V, that would account for the voltage differential on the output. I've attached the schematic with some voltages I measured.

C8 was about the explode, before I resoldered the heatsink transistors there was 63V across the speaker OUT. After that, it dropped to about 25V and I replaced C8.

The bias adjustment does change the DC coming out, but even at its extremes it doesn't help enough. Also, I did remove Q1 and Q3 to measure for shorts, they seem fine.

Thanks
-kdawg

[Enzo]

Hi and welcome. I would really recommend you start a new thread for your amp instead of adding on the end of an old one. More people will notice it.

ANyway, if the outputs Q1,3 were shorted, you'd blow fuses. You have DC on the output, as you know. Your output stage appears to be stable, but driven over to one side. If there is +20v on the driver bases, it will be on the output as well. The problem is farther back.

By the way, always work without a load on the amp until you cure the DC offset.

Looking at the drawing, it appears that clockwise is the coldest bias setting, so set it there. If I am wrong, then set it for coldest - least mains draw.

Q10 provides current for the bias string and the rest of the front end of the amp. Q6 is your voltage amp stage - some refer to it as the VAS. By controling the curent through the bias string, it makes the driver bases go up and down to the music. ANd thus they send the outputs the same way. IF Q6 doesn;t cunduct enough, then the bias string rises towards Q10 and the +rail. 20v in your case.

I'd see if Q6 was OK, myself. Frankly on an amp this simple ther are only a few transistors. I'd check them all with your meter on diode test.

SInce C8 was damaged, it is entirely possible the differential pair - Q7,8 - was damaged. Solid state amps are very heavily fed back. Q8 controls your VAS, but it is up to Q7 to monitor the output and correct for any errors. If Q7 doesn;t work, then it cannot compensate for any offset leanings in Q8 and the VAS. Or ti may be trying to hard to compensate for something.

[kdawg]

Hello Enzo,

Thanks for the info. I tested all transistors and couldn't find anything. I had already replaced Q7, Q8 with NTE234 but that may be the problem, I think that is incorrect substitution. I will try and find right sub - any suggestions?

[Enzo]

2sa1268?